• Swine Flu: waiting for data

    by  • April 27, 2009 • Flu • 0 Comments

    So here’s the basic situation as I understand it, as of Monday morning, April 27th London time. It’s mainly defined by questions, as you might expect this early in the situation. Here are the questions and the data we have so far.

    1> What’s the CFR?
    The lethality of a flu is known as its CFR (case fatality rate): the odds of dying if one is infected.

    We have poor data from Mexico. Something like 1600 cases, something like 100 fatalities. Rumors of many more fatalities circulate, but there is also a rational expectation that many milder cases went undetected. The data from Mexico gives us a near certain upper bound on the CFR of the flu: about 7%. It’s extremely unlikely to be worse than that even when all the data is in.

    The lower bound, however is unknown at this point. I haven’t even seen initial estimates from any major source. So we’re guessing. Broadly speaking the terrain in terms of public health response is similar from maybe 0.5% to 2% or 3%: the final death tolls would be very different, but the preventative measures are fairly similar.

    2> What’s the CAR?
    The infectiousness of a flu is known as its CAR (case attack rate): the percentage of the population that gets infected. Death toll is CAR * CFR * Population size. This number gets very large very quickly.

    There are two factors which affect the CAR of an organism: its own biology, and its environment. We want to be bad hosts: masks, hand washing, keeping away from crowds (“social distancing” as it is called in the trade) to keep the CAR down as far as possible. At this point we don’t know much about the nature of the bug – just how infectious is it? But we do know how flu spreads in general – airborne droplets of moisture from sneezes and coughs, mostly – so we know how to be bad hosts. (see follow the flu code)

    We’re in a situation here where we don’t exactly understand our opponent, the flu, but we do understand ourselves and how to stay away from our opponent, and that is a pretty good start. We have partial control of the CAR. We should use it.

    3> How widespread is it?
    Scary maps aside so far we don’t have major outbreaks outside of Mexico. So far everybody infected that I’ve read about had been to Mexico or was in extremely close proximity for prolonged periods with people who had. We’re not seeing – so far – reports of somebody sneezing on a train and infecting 80 people.

    Now… is that because it has a long incubation time (very, very bad) or because it’s just not that infectious (very, very good) or just because we don’t have data yet? (most probable)

    The fundamental deal
    Behavior so far, interpolated from available data, is perfectly consistent with an outbreak that could kill a hundred million people. It’s also consistent with a much milder situation. The critical question is “why are the reports from Mexico so much worse than those from the USA?” and, right now, nobody knows.

    What we do know is that we can control the CAR by changing our behavior including doing things like simply staying at home for weeks or months while major outbreaks pass by. If everybody in Mexico could go home, right now, for a month the odds are they would re-contain the flu. People who are sick would not infect very many if any other people, and the infection would slow down or die out. We can deploy this kind of approach to any infectious organism, but it is particularly important for pandemic flu threats.

    The hard part is getting people to do it. So what you can do, for yourself and for others, is to do your bit to lower the CAR of any flu that gets lose in your country, town or city. Get ready to go home, stay home, and wait. Get ready to use masks and very, very frequent hand washing. These things, done effectively, done as a society, are probably as effective as a vaccine at stopping pandemic diseases. But to get that level of protection, everybody has to work together to deny the virus any place to replicate. By keeping safe, we keep our friends safe, and if everybody does that, very, very few people will get ill.

    The internet is an ideal platform for massive behavioral change. It is our collective immune system. Get ready.

    If the swine flu or bird flu goes fully pandemic, rapid response as outlined in the Pandemic Flu / Swine Flu Orientation and Action Guide is the kind of stuff that can save the lives of you, your family and your friends. Collective action can defeat these organisms, and we need to be prepared to take it.

    flattr this!


    Vinay Gupta is a consultant on disaster relief and risk management.


    Leave a Reply

    Your email address will not be published. Required fields are marked *